电信科学 ›› 2018, Vol. 34 ›› Issue (9): 143-152.doi: 10.11959/j.issn.1000-0801.2018243

• 运营技术广角 • 上一篇    下一篇

移动互联网智能管道自组织架构研究与实践

薛飞1,2,徐宇辉1,张靓1,李英1,3   

  1. 1 中国移动通信集团广东有限公司,广东 广州 510623
    2 中山大学岭南(大学)学院,广东 广州 510275
    3 西安电子科技大学电子工程学院,陕西 西安 710071
  • 出版日期:2018-09-20 发布日期:2018-10-12

Research and practice on self-organization architecture of smart pipe in mobile internet

Fei XUE1,2,Yuhui XU1,Liang ZHANG1,Ying LI1,3   

  1. 1 China Mobile Group Guangdong Co.,Ltd.,Guangzhou 510623,China
    2 Lingnan (University) College,Sun Yat-sen University,Guangzhou 510275,China
    3 School of Electronic Engineering,Xidian University,Xi’an 710071,China
  • Online:2018-09-20 Published:2018-10-12

摘要:

根据 3GPP 自组织网络规范,提出移动互联网智能管道自组织架构。在智能管道网络中引入集中策略控制功能实体,周期监测外部消息,利用循环神经网络构建最佳策略匹配模型,进行智能管道策略预测、迭代更新和策略下发。对比传统智能管道,本架构实现了智能管道策略智能化定制、实时性管理以及生命周期动态管理,并在“大流量用户”限速、网络高负荷保障以及在线广告推荐等场景得到了有效的应用。

关键词: 智能管道, 集中策略控制功能, 自组织网络, 循环神经网络

Abstract:

According to the 3GPP specification,a new self-organization network of smart pipe in mobile internet was addressed.By introducing integrated policy control function,the external signals were supervised periodically,and the best optimized network strategies were predicted by the recurrent neural network algorithm,delivered by PCRF,and updated automatically with feedback of the regulations.This solution was effectively applied in scenarios such as speed limit of large traffic users,network protection under high load,and online advertisement recommendation etc.

Key words: smart pipe, integrated policy control function, self-organization network, recurrent neural network

No Suggested Reading articles found!